Intelligent workload balance control of the assembly process considering condition-based maintenance

Abstract

Balancing the workloads of workstations is key to the efficiency of an assembly line. However, the initial balance can be broken by the changing processing abilities of machines because of machine degradation, and at some point, re-balancing of the line is inevitable. Nevertheless, the impacts of unexpected events on assembly line re-balancing are always ignored. With the advanced sensor technologies and Internet of Things, the machine degradation process can be monitored continuously, and condition-based maintenance can be implemented to improve the health state of each machine. With the technology of robotic process automation, workflows of the assembly process can be smoothed and workstations can work autonomously together. A higher level of process automation can be achieved. Real-time information of the processing abilities of machines will bring new opportunities for automated workload balance via adaptive decision-making. In this study, a fuzzy control system is developed to make real-time decisions to balance the workloads based on the processing abilities of workstations, given the policy of condition-based maintenance. Fuzzy controllers are used to decide whether to re-balance the assembly line and how to adjust the production rate of each workstation. The numerical experiments show that the buffer level of the assembly line with the proposed fuzzy control system is lower than that of the assembly line without any control system and the buffer level of the assembly line with another control system is the lowest. The demand can always be satisfied by assembly lines except the one with another control system since there is too much production loss sacrificed for the low buffer level. The sensitivity analysis of the control performance to the parameter settings is also conducted. Thus, the effectiveness of the proposed fuzzy control system is demonstrated, and intelligent automation can improve the performance of the assembly process by the fuzzy control system since real-time information of the assembly line can be used for adaptive decision-making.

Introduction

Assembly line balancing problem (ALBP) is key to the efficiency of the assembly process. However, the assembly process is prone to disruptions. Disruptions associated with workstations, buffers or raw materials can affect the assembly process negatively, and the adverse impact can propagate along the assembly line. Thus, the assembly process should be monitored and controlled in real time to smooth the workflow. Industry 4.0, the fourth industrial revolution, brings new possibilities for the assembly systems. Internet of Things (IoT), as the basic premise for the implementation of Industry 4.0 [1], has a significant influence on the developments of smart workshop [2]. Cyber-Physical Systems (CPS), which is another key technology of Industry 4.0, can manage the integration of data collected from a factory, and enables information to be monitored and synchronized between the physical world and the cyber environment [3]. With the real-time information obtained, the production process can be optimized and re-optimized, thus, intelligent manufacturing, which will bring revolutionary changes [4], is possible. Smart systems such as smart robotic mobile fulfillment system [5], warehouse management system for smart logistics [6], smart product-service systems [7], [8], smart active maintenance system [4], smart suite for smart factory towards re-industrialization [9] and smart home system [10] have been proposed based on the advanced technologies of Industry 4.0.

With the development of information and communication technologies (ICT), information technology is embedded in normal products and transforms them to smart, connected products so that information can be generated [11]. IoT brings internet to all kinds of devices so that they can not only collect and send information but also receive information. Additionally, IoT can significantly affect the tracking applications [2], therefore, the work-in-progress during the production process can be tracked more easily and accurately. Due to the digital transformation enabled by ICT, physical products can be digitized in the virtual space and interconnected [12]. As a result, a workstation of an assembly line can ‘communicate’ with its upstream and downstream workstations and buffers to get access to both local information and global information. Then, information barriers between different parts of an assembly line can be broken, and centralized and decentralized decision-making are possible. Robotic process automation (RPA), a new technology that consists of software agents that mimic the manual routine decisions via various computer applications, has attracted attention in Industry and academia [13]. It is enabled by the advanced information technologies, and brings new possibilities to the assembly process: workstations can work autonomously with each other to satisfy the demand, and can collaborate with each other to deal with unexpected events and decrease work-in-progress. Since the benefits brought by the advanced technologies can be fused by the automation with analytics and decision-making via tools of artificial intelligence, traditional automation of the assembly process can be improved to be intelligent automation [14]. Therefore, an intelligent control system of the assembly process enabled by the latest technologies of Industry 4.0 is possible, and it is necessary to decrease the negative impacts of unexpected events and increase the line efficiency.

Machine degradation can affect the duration of job processing, and will negatively affect the workload balance of an assembly line. Thus, the degradation process and machine state should be monitored so that workloads of workstations can be balanced with the real-time processing abilities of workstations. With the development of sensor technologies, it is possible to monitor the machine degradation process, and the information collected can be used to build and optimize the schedule of condition-based maintenance, which is to implement maintenance operations based on the online measurements of machine degradation level [15]. Besides, IoT enables real-time information collection to be possible, and helps establish efficient maintenance strategies at low cost with the related information [10]. The findings of Ghaleb et al. [16] indicated that significantly increase savings could be obtained when the accurate information of machines’ degradation process were incorporated in the condition-based maintenance strategies. Thus, smart machines can send their working status to a central cloud-based “manager” [17], and real-time degradation information can be levered to implement condition-based maintenance and improve the health state of each machine at the right time.

Some researchers explored the production scheduling problems and modeled the impact of deterioration effect on job processing time with a linear or non-linear function of start time (e.g. [18], [19]) or the position of the job (e.g. [20], [21]). However, the link between the extent of machine degradation and production re-scheduling was not discussed by these studies. With respect to assembly line re-balancing, to react quickly to the disruptions to the assembly process, Belassiria et al. [22] and Girit and Azizoğlu [23] explored the assembly line re-balancing solution with methods based on genetic algorithm and both exact and tabu search algorithm, respectively. Meanwhile, to respond quickly to changes, Moghaddam et al. [24] utilized reconfigurable manufacturing systems with modular reconfigurable machine tools to adjust production capacity of the system. However, these researches focused on searching the re-optimization solutions with the assumption that the re-balancing decision has been made already, and when to re-balance the assembly line was still not examined.

The link between the extent of disruptions and the decision-making process of assembly line re-balancing is examined by only a few researches. Suwa [25] and Valledor et al. [26] proposed the periodic re-scheduling policy, and whether to implement re-scheduling was determined based on the condition at the predetermine inspection times. However, the effectiveness of such policies would inevitably depend on the predetermined inspection interval, and the real-time decision-making was not possible. In our previous work [27], [28], a fuzzy control system was proposed to determine when to re-balance the workloads and manage the inventory level of work-in-progress. Nevertheless, it was assumed that no maintenance activities would be conducted until machines of an assembly line broke down. Therefore, it is necessary to build a real-time control system and develop a link between the extent of disruptions and the re-balancing decision, and real-time information of the machine degradation process should be utilized to implement condition-based maintenance.

Due to the randomness and non-linearity caused by unpredictable disruptions, accurate analysis of the assembly line is difficult. However, fuzzy controllers provide an efficient architecture to incorporate the linguistic information from the knowledge of experts to the final automated decisions. Thus, fuzzy control theory is chosen to design the control system of the assembly process, and is used to determine the trigger point of assembly line re-balancing in a novel way. To fill the above research gaps, this study integrates the fuzzy logic principles and the assembly line monitoring and control mechanism, and enables the effective decision-making process of how to manage the inventory level of the work-in-progress and when to re-balance the assembly line, which is of great importance to the efficiency of assembly lines but always ignored in the current literature. During the control process, fluctuations of processing abilities of machines due to machine degradation and condition-based maintenance are considered. A fuzzy control system is proposed to analyze the real-time information collected from the assembly process and balance the workloads by not only adjusting the production rates of workstations but also re-balancing the workloads by task re-assignments when necessary. Two kinds of assembly line re-balancing strategies are used so that two levels of modifications of the initial assembly plan can be implemented to the assembly line. The machine degradation process is assumed to be monitored continuously in real time so that the changing processing abilities of workstations can be considered in the decision-making process. This study explores the smart balance control of an assembly line with the consideration of the information brought by the advanced technologies in Industry 4.0. The research findings will shed light on the intelligent control of the assembly process and contribute to the smart manufacturing theory.

The remaining sections of this paper are organized as follows. In Section 2, literature on production scheduling considering the deterioration effect and the real-time control of the production process is reviewed. Section 3 shows the problem definition and the proposed fuzzy control system. The results of the numerical experiments are shown in Section 4. Section 5 presents the conclusions of this study.